isolates that show resistance to methicillin (methicillin-resistant S. aureus
, MRSA) were first reported in the United Kingdom in 1961, and soon after, in other European countries, Japan, Australia, and the US [1
]. According to the Centers for Disease Control and Prevention (CDC), in the US, there are more than 35,000 deaths and 90,000 severe cases of MRSA infection each year [2
]. Vancomycin is typically the antibiotic of choice for treating serious Gram-positive bacterial infections, including MRSA, but reports of vancomycin-resistant and vancomycin-intermediate S. aureus
strains are becoming more common [3
Whole animal-based high-throughput screening (HTS) is a powerful tool for discovering antibacterial compounds, including those that would otherwise be missed by traditional antibiotic screens that report compounds that inhibit bacterial growth [4
]. Our laboratory has developed a fully automated HTS platform that utilizes the nematode Caenorhabditis elegans
to identify compounds that cure Staphylococcus aureus
], Enterococcus faecalis
], or Pseudomonas aeruginosa
] infections in C. elegans
without showing frank host toxicity that results in the death of the nematodes. Since these pathogens kill C. elegans
, hit compounds are those that reduce C. elegans
killing, resulting in an excess of live worms compared to non-treated pathogen-infected nematodes. Highly toxic compounds do not emerge as hits in the assay, because they kill the nematodes.
Here, we describe the coupling of the C. elegans
HTS platform with a secondary screen that uses MRSA infection of Galleria mellonella
(wax moth) larvae (caterpillars) to validate the anti-staphylococcal activity and low whole-animal toxicity of hit compounds at the whole-organism level [8
]. In this study, we used the dual-host approach to identify and validate five new anti-MRSA compounds, and report follow-up characterization of their antibacterial and other properties.
The aim of this study was to characterize the antibacterial properties of five synthetic compounds with activity against MRSA. The activity of the compounds against MRSA was originally identified using a C. elegans–
MRSA whole-animal infection model. In this study, compound activities were confirmed at the whole-animal level using the G. mellonella
–MRSA model [5
]. The G. mellonella
infection model is a quick and inexpensive way to confirm the efficacy of hit compounds against S. aureus
in vivo [17
], but is too cumbersome and labor intensive to implement for large scale screening of chemical libraries. Hence, we used the G. mellonella
model for follow-up validation of C. elegans
high throughput screening hits to enable prioritization of compounds for further characterization.
The five compounds chosen for study here (PPT, NNC, TBB, GW4064, and PD198306) all prolonged the survival of both C. elegans
and G. mellonella
infected with MRSA. A literature survey revealed that although the compounds had reported bioactivities, none had previously been reported to have antibacterial activity. PPT is described as a subtype-selective ERα agonist. We found that the MIC of PPT against MRSA-MW2 and four clinical isolates was 8 µg/mL. The compound showed potent membrane permeabilizing properties and some hemolytic activity in human red blood cells, suggesting it targets the bacterial membrane. Synergy was observed with PPT and the antibiotic doxycycline. As a class, pyrazoles show a wide range of biological activities, including antimicrobial, antifungal, antitubercular, anti-inflammatory, anticancer, antiviral, neuroprotective, and estrogen receptor (ER) activity [10
]. Some pyrazole derivatives have been shown to exhibit activity against both Gram positive and Gram-negative bacteria and an N
-pyrazole-4-carboxamide derivative increased the survival of G. mellonella
infected with S. aureus
]. PPT is a highly hydrophobic pyrazole containing three phenolic substituents and it is possible that these properties are responsible for the significant toxicity observed against eukaryotic cells.
NNC is a reported calcium channel blocker [11
]. The compound showed a MIC of 8 µg/mL against MRSA MW2 and was slightly more potent against the other four isolates. As with PPT, NNC appears to target the bacterial membrane, since it permeabilizes MRSA cells and shows hemolytic activity at 32 µg/mL. It was also found to be cytotoxic against all three eukaryotic cell lines.
TBB is a reported inhibitor of protein kinase 2 (CK2) [12
]. The compound showed a MIC between 4–6 µg/mL against MRSA and the clinical isolates and while it appears to be membrane active against MRSA-MW2, no red blood cell hemolysis was observed. The compound synergized with doxycycline and oxacillin against MRSA, however, like PPT and NNC, it was also toxic to eukaryotic cell lines at concentrations around its MIC.
GW4064 is a reported farnesoid X receptor (FXR) agonist [19
] that suppresses cell proliferation in several cancer lines, but not normal cells [20
]. The compound showed a MIC of 4–8 µg/mL against the MRSA strains and synergized with gentamycin. While GW4064 appears not to be membrane active against MRSA MW2, it showed hemolytic activity at 8 µg/mL and was cytotoxic to eukaryotic cells.
PD198306 is a potent and selective non-ATP competitive inhibitor of MEK1/2 that shows antihyperalgesic properties [14
]. The compound showed the lowest MIC (2 µg/mL) against MRSA MW2, which was similar for the four other isolates. Its activity was bacteriostatic and the compound synergized with erythromycin. Importantly, PD198306 does not appear to be membrane active against MRSA and shows no hemolytic activity, suggesting that its antibacterial action may be more target-specific than the other compounds. However, PD198306 was also cytotoxic towards the three eukaryotic cell lines at concentrations around its MIC, potentially reducing its attractiveness for further study.
In summary, this study demonstrates the utility of the C. elegans/G. mellonella whole animal dual-screening approach for identifying compounds of interest for study as new agents against MRSA. The sequential use of two independent model hosts is a powerful tool for prioritizing compounds, as it confirms that efficacious anti-microbial compounds identified by HTS in C. elegans are indeed relatively non-toxic at a whole animal level and may warrant further study in vertebrate animals. Although none of the five compounds exhibited a high level of toxicity at the whole animal level, all five compounds did inhibit the proliferation of human cells at concentrations approximating the MICs for MRSA. Despite this, the chemical scaffolds, particularly PD198306, represent good starting points for traditional medicinal chemistry aimed at decreasing toxicity while maintaining antimicrobial activity.